Back to Frameworks
TPL-18Reliability

Architecture Decision Record (ADR)

Two years from now, someone will ask 'why did we build it this way?' — and everyone who was in the room will have left. ADRs are the cheapest institutional memory you'll ever buy.

Context

What is the situation and what forces are at play? Business constraints, technical constraints, timing. Enough that a new joiner could understand why this was even a decision.

Options considered

At least two, ideally three. Each with a short pros/cons. If you can only name one option, you didn't make a decision.

Decision

Which option, in a single crisp sentence. Include the date and the deciders.

Consequences

What becomes easier, what becomes harder, and what we're accepting as risk. This is the section future-you will thank present-you for writing.

Common pitfalls

  • Writing ADRs for trivial choices. Reserve them for decisions with consequences.
  • Skipping 'options considered.' Kills the value.
  • Editing an ADR after the fact. Supersede it with a new one instead.

Leadership in your inbox.

Leadership lessons, frameworks, and field notes for modern engineering teams. No motivational fluff.

Trouble seeing the form?Subscribe on Beehiiv instead

Free. One email a week. Unsubscribe anytime.